Skip to content
Markdown by mdedit.ai

Give your AI agent a real Markdown workspace

One plugin brings the mdedit connection and five skills, Save, Find, Revise, Review, and Publish, into ChatGPT, Claude Code, Cursor, Gemini CLI, and VS Code. Agents draft and suggest; you approve what lands.

By Vivek Maskara · Last updated

You

Review my launch brief and leave suggestions for anything unclear.

Agent

3 suggestions added, nothing changed in your text.

Suggestion 1 of 3

We plan to ship the update soon to most customers.

The update reaches every paid workspace on 14 October.

Suggestions wait for you in the editor

What you'd ask for

Four prompts you can paste into a connected client, and what each one does to your document.

Review this draft and leave suggestions instead of editing directly.

The agent anchors suggestions to exact passages. You accept or reject each one in the editor.

Summarize the open review threads and resolve the ones I already replied to.

Comments, replies, and resolutions land in the same review panel your team uses.

Create a document from these meeting notes in my Docs workspace.

A durable document appears in your workspace, ready to edit together.

Publish this doc as a live link.

The agent must ask first: publishing requires an explicit confirmation and its own key scope. The mded.it link then follows future edits.

Install for your host

The same plugin, version 0.3.1, installs four ways. Each host handles its own sign-in.

ChatGPT and Codex

Available

Install Markdown by mdedit.ai from the ChatGPT Store.

Claude Code

Awaiting marketplace listing

Add the MCP connection with one command, or load the plugin from the shared repository.

terminal
claude mcp add --transport stdio mdedit -- npx -y @mdedit/mcp-server

The plugin is ready and awaiting marketplace listing. Until then, load it from the shared repository or use CLI-backed skills.

Cursor and VS Code

Awaiting marketplace listing

In VS Code, run Chat: Install Plugin From Source with the repository URL. In Cursor, install the same repository from source.

The plugin is ready and awaiting marketplace listing. Until then, install the shared repository from source.

Gemini CLI

Awaiting marketplace listing

Install the extension with the repository URL, then authenticate when prompted.

terminal
gemini extensions install \
  https://github.com/mdedit-ai/mdedit-agent-plugin

The extension is ready and awaiting gallery listing. Until then, install it with the repository URL.

Skills-only installs and every host's details are in Install mdedit agent skills.

The agent proposes, you approve

Suggestions wait for a person

An agent proposal stays anchored to its passage until you accept or reject it in the editor.

Publishing asks first

A public link needs the publishing:write scope and an explicit confirmation on the tool call.

Nothing beyond the grant

Every action is limited to what the key or the sign-in allows, and anything else fails closed.

The five skills tell the agent when to save, find, revise, review, or publish, so it reaches for the right tool at the right moment. Read the skill catalog.

Connect a custom MCP client

Create a scoped API key

Works with every MCP client. Create the key in Settings, Integrations, API Keys.

terminal
# Pass a scoped key from Settings
claude mcp add --transport stdio \
  --env MDEDIT_API_KEY=mdh_your_key \
  mdedit -- npx -y @mdedit/mcp-server

Hosted endpoint for HTTP clients

POST https://mcp.mdedit.ai/mcp
X-API-Key: mdh_your_key

Clients that browse the official MCP registry can find the server as io.github.mangoappstudio/mdedit.

Already use the mdedit.ai CLI?

Run mdedit auth login once and the local MCP server picks up that sign-in from your operating system keychain, so nothing secret goes into your client config. Browser sign-in is rolling out gradually and may not be enabled for every account yet. Node.js 20 or newer is required for the local package.

Client by client setup lives in the Claude Code guide, and connection troubleshooting in MCP authentication.

Tools at a glance

Read and explore

4 tools

Find the right workspace and open a document before touching anything.

Write, review and suggest

9 tools

Draft, propose, and work through feedback where your team already reads it.

Export and convert

4 tools

Turn a document into Markdown, HTML, PDF, DOCX or a Zip bundle, and bring outside files in.

Publish and presence

5 tools

Public links stay behind a confirmation, and agents appear alongside your teammates.

Twenty-two tools total. Full list in the agent docs.

One key per agent, scoped to one job

Keys start with mdh_ and carry a fixed set of permissions.

Review only

articles:read, reviews:write

Reads documents and leaves comments and suggestions. It cannot rewrite your prose.

Editor

articles:write

Creates documents and applies anchored edits. A write scope also covers the matching read.

Publisher

publishing:write

Creates or disables a public mded.it link, and only with the confirmation flag set.

A tool the key cannot reach returns a permission challenge instead of doing half the work. Read the API key guide.

Other ways to automate mdedit.ai

Command line

Script workspaces, documents, and exports from your terminal, with JSON output for pipelines.

See the CLI

Webhooks

Wake your own process when a reviewer mentions an agent or a document changes.

Read the webhook guide

Agent plugin FAQs

One plugin, in the Agent Plugins 1.0 format. It bundles the hosted mdedit MCP connection with five skills: Save, Find, Revise, Review, and Publish. The same package carries thin host adapters for Claude Code, Cursor, and Gemini CLI, and the plugin is listed in the ChatGPT Store as Markdown by mdedit.ai. See installation by host.

The Model Context Protocol is an open standard for connecting AI clients to external tools. An MCP server publishes a list of typed tools, and the client decides which ones to call. @mdedit/mcp-server is the mdedit.ai implementation, and it publishes twenty-two tools for Markdown documents.

Often, yes. Hosts that install the plugin handle sign-in themselves, with browser OAuth where that flow is marked verified. Locally, mdedit auth login shares its operating system keychain sign-in with the MCP server, though browser sign-in is rolling out gradually and may not be enabled for your account yet. A scoped API key works everywhere else. See MCP authentication.

Anything that speaks MCP over standard input and output can run the local package, either with the shared CLI sign-in or with MDEDIT_API_KEY in its environment. Claude Code has a documented setup command. Clients that support Streamable HTTP and custom headers can use the hosted endpoint with an API key instead.

Run the local package when your client launches processes, which is the usual case for terminal agents. Use https://mcp.mdedit.ai/mcp with an X-API-Key header when your client only speaks HTTP. Both paths expose the same twenty-two tools. See MCP authentication.

No. Publishing needs the publishing:write scope, and publish_article also needs an explicit confirmation flag on the call. An ordinary editing key cannot make a document public. Once a link exists, anyone who has it can read the document, so treat that step as a deliberate one.

The call fails closed. The tool returns a permission challenge naming the missing scope instead of partially completing the work, so a review-only key stays review-only.

No. add_suggestion records a proposed replacement against a passage, and it stays a proposal until a person accepts it in the editor. add_comment never changes prose at all. Only edit_article writes to the document, and it needs a grant scoped for writing.

They serve different jobs, and they share one sign-in. MCP is for an agent working inside a document during a session. The mdedit.ai CLI is for scripts, pipelines, and terminal work where you want plain commands and JSON output.

Ready to connect

Give your agent a document to work in

Install the plugin for your host, and keep the final say on every change.